beagled/Lucene.Net/Analysis/StopFilter.cs

   1 /*
   2  * Copyright 2004 The Apache Software Foundation
   3  *
   4  * Licensed under the Apache License, Version 2.0 (the "License");
   5  * you may not use this file except in compliance with the License.
   6  * You may obtain a copy of the License at
   7  *
   8  * http://www.apache.org/licenses/LICENSE-2.0
   9  *
  10  * Unless required by applicable law or agreed to in writing, software
  11  * distributed under the License is distributed on an "AS IS" BASIS,
  12  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  13  * See the License for the specific language governing permissions and
  14  * limitations under the License.
  15  */
  16 using System;
  17 namespace Lucene.Net.Analysis
  18 {
  19
  20         /// <summary> Removes stop words from a token stream.</summary>
  21
  22         public sealed class StopFilter : TokenFilter
  23         {
  24
  25                 private System.Collections.Hashtable stopWords;
  26
  27                 /// <summary> Constructs a filter which removes words from the input
  28                 /// TokenStream that are named in the array of words.
  29                 /// </summary>
  30                 public StopFilter(TokenStream in_Renamed, System.String[] stopWords) : base(in_Renamed)
  31                 {
  32                         this.stopWords = MakeStopSet(stopWords);
  33                 }
  34
  35                 /// <summary> Constructs a filter which removes words from the input
  36                 /// TokenStream that are named in the Hashtable.
  37                 ///
  38                 /// </summary>
  39                 /// <deprecated> Use {@link #StopFilter(TokenStream, Set)} instead
  40                 /// </deprecated>
  41                 public StopFilter(TokenStream in_Renamed, System.Collections.Hashtable stopTable) : base(in_Renamed)
  42                 {
  43                         stopWords = new System.Collections.Hashtable(new System.Collections.Hashtable(stopTable));
  44                 }
  45
  46                 /// <summary> Builds a Hashtable from an array of stop words,
  47                 /// appropriate for passing into the StopFilter constructor.
  48                 /// This permits this table construction to be cached once when
  49                 /// an Analyzer is constructed.
  50                 ///
  51                 /// </summary>
  52                 /// <deprecated> Use {@link #MakeStopSet(String[])} instead.
  53                 /// </deprecated>
  54                 public static System.Collections.Hashtable MakeStopTable(System.String[] stopWords)
  55                 {
  56                         System.Collections.Hashtable stopTable = System.Collections.Hashtable.Synchronized(new System.Collections.Hashtable(stopWords.Length));
  57                         for (int i = 0; i < stopWords.Length; i++)
  58                                 stopTable[stopWords[i]] = stopWords[i];
  59                         return stopTable;
  60                 }
  61
  62                 /// <summary> Builds a Set from an array of stop words,
  63                 /// appropriate for passing into the StopFilter constructor.
  64                 /// This permits this stopWords construction to be cached once when
  65                 /// an Analyzer is constructed.
  66                 /// </summary>
  67                 public static System.Collections.Hashtable MakeStopSet(System.String[] stopWords)
  68                 {
  69                         System.Collections.Hashtable stopTable = new System.Collections.Hashtable(stopWords.Length);
  70                         for (int i = 0; i < stopWords.Length; i++)
  71                                 stopTable.Add(stopWords[i], stopWords[i]);
  72                         return stopTable;
  73                 }
  74
  75                 /// <summary> Returns the next input Token whose termText() is not a stop word.</summary>
  76                 public override Token Next()
  77                 {
  78                         // return the first non-stop word found
  79                         for (Token token = input.Next(); token != null; token = input.Next())
  80                                 if (!stopWords.Contains(token.termText))
  81                                         return token;
  82                         // reached EOS -- return null
  83                         return null;
  84                 }
  85         }
  86 }